Tilburg University
Exploring the personality structure in the 11 languages of South Africa
Nel, J.A.; Valchev, V.H.; Rothmans, S.; van de Vijver, Fons; Meiring, D.; de Bruin, G.P.
Published in:
Journal of Personality
Document version:
Publisher's PDF, also known as Version of record
DOI:
10.1111/j.1467-6494.2011.00751.x
Publication date:
2012
Link to publication
Citation for published version (APA):
Nel, J. A., Valchev, V. H., Rothmans, S., van de Vijver, F. J. R., Meiring, D., & de Bruin, G. P. (2012). Exploring
the personality structure in the 11 languages of South Africa. Journal of Personality, 80, 915-948. DOI:
10.1111/j.1467-6494.2011.00751.x
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners
and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
- Users may download and print one copy of any publication from the public portal for the purpose of private study or research
- You may not further distribute the material or use it for any profit-making activity or commercial gain
- You may freely distribute the URL identifying the publication in the public portal
Take down policy
If you believe that this document breaches copyright, please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
Download date: 15. jun. 2017
Exploring the Personality Structure in the 11
Languages of South Africa
Jan Alewyn Nel,1 Velichko H. Valchev,2
Sebastiaan Rothmann,1 Fons J. R. van de Vijver,1,2
Deon Meiring,3 and Gideon P. de Bruin4
1
2
North-West University, South Africa
Tilburg University, The Netherlands
3
University of Pretoria, South Africa
4
University of Johannesburg, South Africa
ABSTRACT The present study, part of the development of the South
African Personality Inventory (SAPI), explores the implicit personality
structure in the 11 official language groups of South Africa by employing
a mixed-method approach. In the first, qualitative part of the study,
semistructured interviews were conducted with 1,216 participants from
the 11 official language groups. The derived personality-descriptive terms
were categorized and clustered based on their semantic relations in iterative steps involving group discussions and contacts with language and
cultural experts. This analysis identified 37 subclusters, which could be
merged in 9 broad clusters: Conscientiousness, Emotional Stability,
Extraversion, Facilitating, Integrity, Intellect, Openness, Relationship
The SAPI, an acronym for the South African Personality Inventory, is a project that
aims to develop an indigenous personality measure for all 11 official languages in South
Africa. Participants are Byron Adams (University of Johannesburg and Tilburg University), Gideon P. de Bruin (University of Johannesburg), Karina de Bruin (University
of Johannesburg), Carin Hill (University of Johannesburg), Leon Jackson (North-West
University), Deon Meiring (University of Pretoria and University of Stellenbosch),
Jan Alewyn Nel (North-West University), Ian Rothmann (North-West University),
Michael Temane (North-West University), Velichko H. Valchev (Tilburg University),
and Fons J. R. van de Vijver (North-West University and Tilburg University).
Special thanks are expressed to Boele de Raad for his comments on the final
versions of the model presented in this study.
Correspondence concerning this article should be addressed to Jan Alewyn Nel,
North-West University, School for Human Resource Sciences, Potchefstroom, NorthWest, 2520, South Africa. Email: [email protected].
Journal of Personality 80:4, August 2012
© 2011 The Authors
Journal of Personality © 2011, Wiley Periodicals, Inc.
DOI: 10.1111/j.1467-6494.2011.00751.x
916
Nel, Valchev, Rothmann, et al.
Harmony, and Soft-Heartedness. In the second, quantitative part, the
perceived relations between the 37 subclusters were rated by 204 students
from different language groups in South Africa and 95 students in the
Netherlands. The outcomes generally supported the adequacy of the conceptual model, although several clusters in the domain of relational
and social functioning did not replicate in detail. The outcomes of these
studies revealed a personality structure with a strong emphasis on socialrelational aspects of personality.
Personality inventories are mostly developed from existing, usually
Western, personality models. Even if it is assumed that personality
structure is universal, there may be cross-cultural variations in the
expression of this structure, which have implications for assessment
(Berry, Poortinga, Segall, & Dasen, 2002). We present the outcomes
of a mixed-method study that explores personality structure in South
Africa, which in the end will be employed to develop a new personality inventory (the South African Personality Inventory [SAPI]). We
first provide a brief introduction to current etic and emic approaches
to the study of personality structure, followed by a description of the
comparative lexical approach (a version of which is adopted in the
present study).
Approaches to the Study of Personality Structure
Different approaches have been employed in the exploration of personality structure and the comparison of personality structures
across cultures. The etic approach, usually employing inventories,
focuses on the cross-cultural universality of traits, whereas the emic
(indigenous) approach investigates traits in a particular culture,
thereby maximizing the suitability of the instrument in the target
cultural context (Church, 2001). It is a strength of the etic approach
that it helps to identify commonalities in personality across cultures,
and a weakness that the focus on commonalities may lead to an
underrepresentation of culturally unique aspects. The strengths and
weaknesses of the emic approach are just the opposite. Therefore,
both approaches are complementary. The cross-cultural comparison
of lexical studies has been suggested as a way of combining etic and
emic approaches (F. M. Cheung, Van de Vijver, & Leong, 2011; Lee
& Ashton, 2008; Saucier, 2009; Saucier & Goldberg, 2001).
A typical representative of the etic approach is the body of
research that has found support for the Five-Factor Model (FFM)—
Personality Structure in South Africa
917
describing personality along the dimensions of Extraversion,
Agreeableness, Conscientiousness, Neuroticism, and Openness to
Experience—across a large number of cultures (e.g., McCrae et al.,
2005). In this tradition, a model developed in the United States has
been replicated using structured personality inventories in Western
and non-Western regions, where many different languages from
various language families are spoken (McCrae & Allik, 2002;
McCrae et al., 2005). Several studies have indicated problems with
the cross-cultural replicability of the Openness dimension; yet the
exact role of culture is not clear because no cultural factor has been
identified that could explain when a good (or bad) replication of
Openness could be expected (Church, 2008).
On the other hand, studies in the emic approach set out to
explore the indigenous personality structure in a given culture. F.
M. Cheung and colleagues (2001) studied personality conceptions in
China, starting with assembling everyday-life person descriptions
from Chinese literature, proverbs, and interviews. The qualitative
findings of this exploration served as input for an indigenous
Chinese questionnaire, the Chinese Personality Assessment Inventory (CPAI, and subsequently CPAI-2). The factor-analytic structure observed with this questionnaire had a fair correspondence
with the FFM; however, Openness was found to be weakly represented, and a new concept, labeled Interpersonal Relatedness, was
identified that could not be subsumed under the FFM. This new
factor involves relational aspects of personality, such as maintaining harmony, avoiding conflict, being flexible to situations, and
saving face. Interestingly, subsequent research with the CPAI replicated the Interpersonal Relatedness dimension with other Asian
(S. F. Cheung, F. M. Cheung, Howard, & Lim, 2006), Chinese
American, and even European American samples (Lin & Church,
2004). This suggests that F. M. Cheung and colleagues’ study, starting from an indigenous perspective, has identified a personality construct that is recognizable beyond the specific context of Chinese
culture, although its salience in other cultures may be different.
Using a similar research approach, Katigbak and colleagues (e.g.,
Katigbak, Church, Guanzon-Lapeña, Carlota, & del Pilar, 2002)
conducted a series of studies of indigenous Filipino personality
structure. The dimensions they identified were largely similar to the
FFM, and culture-specific elements were only found for some items
of Broad-Mindedness (Openness).
918
Nel, Valchev, Rothmann, et al.
Lexical Models Across Cultures
The psycholexical method is widely employed in personality research.
It is based on the assumption that salient individual differences in
psychological functioning are embedded or encoded in language
(Allport & Odbert, 1936; Saucier & Goldberg, 2001). Individual
differences that are seen as more prominent are more likely to have
been converted to single words to help describe people. To study
the implicit personality conceptions, personality-descriptive terms
are sampled from dictionaries and research participants are asked to
rate themselves or a familiar other on each term contained in a list.
These ratings are subsequently factor-analyzed. Most lexical studies
report support for the Big Five structure of personality constructs,
closely corresponding to the FFM (Saucier & Goldberg, 2001). An
extensive overview by De Raad et al. (2010), however, suggested that
Extraversion, Agreeableness, and Conscientiousness are the only
factors that fully replicate across languages.
Lexical studies typically employ single person-descriptive terms
extracted from lexica. An alternative approach is to conduct interviews and analyze the generated descriptions, which usually involve
whole phrases in context. Saucier and Goldberg (2001) indicated that
the implicit structure of personality descriptions in phrases or sentences is closely related to that based on single words like nouns
or adjectives. Analysis of free descriptions derived from interviews
has been applied in studies of adult personality (e.g., John, 1990)
and parents’ perceptions of child personality (Harkness et al., 2006;
Kohnstamm, Halverson, Mervielde, & Havill, 1998). The contextual
information found in free descriptions in interviews makes them well
suited for the exploration of indigenous personality conceptions in
different cultures (Mervielde, 1998), which may be especially relevant
if a language uses relatively few abstract trait terms.
To summarize, despite the substantial evidence for universality of
the Big Five model of personality traits coming from studies with
structured inventories (e.g., McCrae & Allik, 2002; McCrae et al.,
2005), lexical studies conducted in different languages have found
less support for universality (e.g., De Raad et al., 2010). In addition,
indigenous studies, notably by F. M. Cheung et al. (2001), have
pointed out that the Big Five model may not be complete, especially
with respect to social aspects of personality. It is evident that indigenous studies in non-Western countries have the potential to detect
Personality Structure in South Africa
919
important personality concepts not well represented by the Big Five
or other Western models. The theoretical debate about the universality of personality dimensions is thus ongoing, and the consensus
on the universality of the Big Five model appears to be weaker than
a few decades ago (Church, 2008; De Raad et al., 2010). While the
most convincing evidence for culture-specific dimensions or additions to the Big Five from a cross-cultural perspective comes from
indigenous studies in China (Church, 2008), it is important to note
that little systematic research has been done on indigenous personality conceptions in Africa.
Personality Study in the South African Context
The general practice in personality research and assessment in South
Africa has been to adopt or adapt tests developed abroad for use in
South Africa (Foxcroft, Paterson, Le Roux, & Herbst, 2004). Most
of these tests did not take into account the political, social, and
economic history of South Africa, and this had a major impact on
psychological assessment for all South Africans. Several studies have
found that these personality inventories showed weak structural
equivalence across ethnic groups and often a low reliability in indigenous African groups (e.g., Abrahams & Mauer, 1999; Meiring, Van
de Vijver, Rothmann, & Barrick, 2005). These psychometric problems could to some extent be attributable to language problems for
populations whose native tongue is one of South Africa’s indigenous
languages; however, item adaptation has not proven a viable way to
solve such problems (Meiring, Van de Vijver, & Rothmann, 2006).
Taylor and De Bruin set out to develop a culturally valid measure
of the FFM in South Africa, taking local context into account. They
found similar factor structures and reliabilities of the five factors of
their Basic Traits Inventory across Black and White groups (Taylor
& De Bruin, 2005) and across different indigenous African language
groups (Ramsay, Taylor, De Bruin, & Meiring, 2008). This work
suggests that personality inventories based on trait models such as
the FFM can yield comparable scores across cultural groups in
South Africa.
Indigenous Concept of Ubuntu
There have been several studies of indigenous African conceptions of
personality (for an overview, see Berry et al., 2002). Although never
920
Nel, Valchev, Rothmann, et al.
worked out in great detail, these models emphasize the relatedness of
persons in groups. An important concept that captures this relatedness is Ubuntu, which is a traditional, everyday notion in South
Africa, especially salient among Black South Africans. Relational
aspects and the social foundation of a person are core in Ubuntu, as
reflected in the Bantu wisdom “A person is only a person through
others.” Ubuntu is associated with social relatedness, peace, and
harmony in a collective- and community-based environment; with
respect for others, tolerance, compassion, and sensitivity toward the
elderly, the handicapped, and the less privileged; with being obedient
toward adults, parents, seniors, and authority; with having courtesy
and loyalty; and with being warm, welcoming, generous, honest, and
trustworthy (Nolte-Schamm, 2006). These elements help in building
and maintaining relationships and are related to the values of collectivism (e.g., Hofstede, 1980). The notion of Ubuntu is also often
quoted as meaning or implying that a person perceives him- or
herself through the perception of others. There are as yet no studies
to substantiate (or refute) the claim that Ubuntu is an indigenous
South African personality construct. Even without such validity
data, the concept of Ubuntu is relevant for our study because it
demonstrates the importance of social and relational aspects of personality in South Africa.
The Present Study
The exploration of the South African personality structure described
in the present study forms part of a bigger project, aimed at creating
an indigenous South African Personality Inventory (SAPI) to overcome current problems facing personality measurement in South
Africa. An important aim of the larger project is the development of
a culturally informed and psychometrically sound instrument to deal
with the rich ethnic and language distribution of the South African
population. There are 11 official languages in South Africa, which
belong to 2 unrelated language families: 2 Germanic (Afrikaans and
English, spoken as a first language by 21.5% of the country’s population) and 9 Bantu languages (Northern Sotho, Southern Sotho,
Tswana, Ndebele, Swati, Xhosa, Zulu, Tsonga, and Venda, spoken
by 77.9% of the population; Statistics South Africa, 2001). Each
language is spoken as a first language by a relatively distinct cultural
group. Germanic-language speakers may belong to one of three
different social-ethnic groups of the apartheid classification still in
Personality Structure in South Africa
921
use today (“White,” 9.6%; “Coloured,” 8.9%; or “Asian/Indian,”
2.5% of the total population), whereas all Bantu-language speakers
are native African (“Black”); English is commonly spoken and
understood by people in all groups.
The present study explores the implicit personality structure as
reflected in the language of speakers of all 11 official languages in
South Africa. Our research relates to the theoretical framework of
the lexical approach. However, we also deviate from it by using
interviews instead of dictionary surveys as means of data generation.
There are three reasons for this. Firstly, dictionaries of a sufficient
quality for our purposes are not available in all official South African
languages. Secondly, some languages do not have many personalitydescriptive terms, which would have led to a potential underrepresentation of relevant concepts. Thirdly, there are few psychologists
available in various language groups who could conduct a lexical
study. Therefore, we adopted the free-descriptions approach and
conducted interviews in which participants were asked to describe in
their native language themselves and particular people they knew
well. Although the lexical approach and our approach have the same
goals (i.e., to identify salient personality descriptors used in a language) and will probably yield similar results, both have their own
strength. The main strength of the lexical approach is its exhaustiveness: A list of personality descriptors based on a dictionary search
finds all relevant terms. The main strength of our approach is ecological validity: Words and expressions found in free descriptions are
actually used in that particular language.
We report two studies. In the first study, we employ a conceptual
analysis of the semantic clustering of personality-descriptive terms
from interviews in all 11 languages. The individual responses obtained
in the interviews are combined in a hierarchical clustering process by
analyzing their semantic relations. The second study attempts to
replicate parts of the clustering process using quantitative methods.
STUDY 1
Method
Participants
Interviews were conducted with participants (N = 1,217) from all 11 language groups. A combination of quota and convenience sampling was
922
Nel, Valchev, Rothmann, et al.
used. The distribution of participants was done in such a manner that
variation was obtained in gender, urban/rural residence, education, and
age. Because speakers of some language groups live mainly in rural areas,
no urban participants were recruited from these groups; sample characteristics are presented in Table 1.
Instrument
Participants were asked to describe themselves and nine other persons
they can be assumed to know well: their best friend of the same sex, their
best friend of the opposite sex, a parent, their eldest child or sibling, a
grandparent, a colleague or friend from another ethnic group, a person
who is the total opposite of the participant, a teacher they liked (if
schooled, otherwise a person from the village whom they liked), and a
teacher they disliked (if schooled, otherwise a person from the village
whom they disliked). In some of the interviews, instead of self-descriptions
and descriptions of a person opposite to oneself, descriptions of a neighbor and of a disliked person were obtained. The following four prompts
were used: “Please describe the following people to me by telling me what
kind of person he or she is/was”; “Can you describe typical aspects of this
person?”; “Can you describe the behavior or habits that are characteristic
of this person?”; and “How would you describe this person to someone
who does not know him/her?” All participants were asked these questions
and there was no limit on the number of characteristic descriptions provided per person.
Procedure
Field-workers who were native speakers of the target language were
recruited and trained to collect data for each of the language groups. The
interviews were conducted in the native language of the participants,
tape-recorded, transcribed, and translated into English by the fieldworkers. Transcriptions were entered in Excel worksheets. Language
experts checked the accuracy of the translations and made corrections
where necessary. There were between 2,300 (Southern Sotho) and 7,300
(English) responses per language group; the total number of responses
was 53,139.
Analysis Outline
In a preparatory stage of the analysis, physical descriptions (e.g., “He has
a dark complexion”), purely evaluative terms (e.g., “He is not good”), and
ambiguous terms (e.g., “She is unlike other girls”) were excluded. This
resulted in the retention of 49,818 responses for the analysis; this number
a
119
43
37
13
16
32.2
(13.2)
30
17
8
10
30.8
(12.3)
55
15
58
61
44
75
English
(n = 119)
44
26
25
45
Afrikaans
(n = 70)
53
67
56
6
31
8
20
3
29.6
(9.2)
62
120
39
21
30
30
35.4
(13.5)
20
42
60
60
S. Sotho
(n = 62)
113
9
43
20
54
5
32
(9.3)
122
58
64
Tswana
(n = 122)
Sotho-Tswana (Bantu)
N. Sotho
(n = 120)
Disputed status; also classified as Sotho-Tswana.
Environment
Urban
Rural
Gender
Male
Female
Ethnic group
Black
Coloured
Indian
White
Age group
16–25
26–35
36–45
46–76
Mean age (SD)
Characteristic
Germanic
101
17
42
28
19
36.2
(12.2)
107
46
61
Ndebelea
(n = 107)
36
79
42
26
22
11
30.9
(11.9)
116
38
69
116
30
36
36
16
34.6
(11.9)
118
51
67
Xhosa
(n = 118)
Nguni (Bantu)
Swati
(n = 116)
Language Group
Table 1
Sample Characteristics per Language
34
107
46
34
49
11
32.5
(9.9)
141
72
69
Zulu
(n = 141)
120
120
33
44
Tsonga
(n = 120)
120
33
48
31
6
31.8
(8.5)
121
68
52
Venda
(n = 121)
Other Bantu
Personality Structure in South Africa
923
924
Nel, Valchev, Rothmann, et al.
includes doubly counted composite responses that were categorized in
more than one category (e.g., the response “cheerful and sociable” was
counted once in each of the respective categories, cheerful and sociable).
The analysis spanned three main stages: labeling, categorizing, and
semantic clustering (for a more detailed description of the analysis
employed on a subset of the data, see Valchev et al., 2011). The general aim
was to reduce the number of statements and categories in an inductive
analysis, based on the semantic similarity and patterns of co-occurrence of
responses, with as few theoretical presumptions as possible. Englishlanguage dictionaries and personality literature were consulted in all stages.
In the labeling stage, we provided common labels for responses with related
but not verbatim identical content (e.g., “He loves going out with friends”
and “He was outgoing” were labeled as outgoing). With this initial grouping
of responses we met two aims: (1) reducing the number of responses to a
more manageable number for further analysis and (2) making labels of
personality-descriptive terms consistent across the language groups. This
stage resulted in over 900 personality-descriptive labels.
In the categorization stage, the responses were categorized in personality facets. We put together synonyms (e.g., outgoing and socializing in
the Sociable facet) and antonyms (e.g., quiet and talkative in the Talkative
facet). This further condensation resulted in a total number of 188 personality facets across languages that represent personality descriptions at
a low-to-medium level of abstraction. Out of the 188 facets, 79 were
extracted in all 11 languages, 71 in 7 to 10 languages, 28 in 3 to 6
languages, and 10 in 1 or 2 languages.
In the semantic clustering stage, we first grouped the personality facets
into more abstract subclusters. The 188 facets were grouped into 37 subclusters based on shared content and patterns of co-occurrence of the
responses (e.g., the Helpful, Supportive, and Community Involvement
facets were assigned to the Active Support subcluster). The analysis aimed
to maximize the homogeneity of personality descriptions within each
subcluster and their heterogeneity across clusters. Finally, the subclusters
were further grouped by means of a conceptual analysis into nine broad
clusters at a level of abstraction similar to that of the Big Five model. The
clusters include two to six subclusters each, and the subclusters include
2 to 12 facets each. The clusters, subclusters, facets, and examples of
constituting responses are presented (alphabetically) in Table 2.
Quality Control
There are no generally agreed-upon procedures for comparative qualitative studies on such a large scale. Therefore, we designed our own checks
and procedures to assess the validity of our inferences. The process
Personality Structure in South Africa
925
Table 2
Clusters, Subclusters, Facets of Personality-Descriptive Terms,
and Example Responses
Cluster
Conscientiousness
Subcluster
Achievement
Orientation
Facet
Example Response (Language)
Career-Oriented (6/27)
She prioritizes—a career before
serious relationship (English)
Likes to compete and compare
herself with other people
(Xhosa)
Dutiful (Afrikaans)
Hard worker (Venda)
Competitive (8/23)
Dedication
Dutiful (1/4)
Hard-Working
(11/1369)
Performance-Oriented
(7/27)
Timeous (1/9)
Dedicated (9/276)
Determined (11/192)
Future-Oriented
(11/145)
Passionate (8/88)
Orderliness
Perseverant (10/261)
Purposeful (5/64)
Consistent (3/38)
Follow-up (2/4)
Meticulous (9/92)
Organized (11/155)
Punctual (11/100)
Self-Discipline
Tidy (10/708)
Thorough (3/21)
Deliberating (9/36)
Disciplined (9/64)
Naughty (11/81)
Obedient (10/110)
Rebellious (8/45)
Serious (8/38)
Thoughtlessness
Absentminded (6/34)
Reckless (9/46)
He likes to achieve everything
by himself (N. Sotho)
Timeous (Afrikaans)
Dedicated to his work (Tswana)
He is determined in everything
he does (S. Sotho)
One who thinks about his
future (Swati)
He does his work
wholeheartedly (Tsonga)
She perseveres (Zulu)
Goal-directed (English)
Consistent (Afrikaans)
She likes to make follow-ups on
things (Ndebele)
Doesn’t have room for mistakes
(Venda)
He is a good planner (Tsonga)
She is always late for her class
(N. Sotho)
Is always clean and tidy (Swati)
Very thorough (English)
He does things without
thinking first (S. Sotho)
He doesn’t have self-discipline
(Tswana)
He is very naughty and doesn’t
listen (Zulu)
Obeys his parents (Venda)
Rebel, dislikes any rules
(Afrikaans)
Serious when time calls for you
to be serious (English)
Is forgetful (Xhosa)
He is careless (Swati)
(Continued)
926
Nel, Valchev, Rothmann, et al.
Table 2 (Cont.)
Cluster
Emotional
Stability
Subcluster
Balance
Courage
Ego
Strength
Emotional
Control
Facet
Balancing Life (2/13)
Even-Tempered
(11/242)
Mature (11/114)
Short-Tempered
(11/660)
Courageous (10/126)
Fearful (11/180)
Attention Seeking
(10/63)
Demanding (10/110)
Needy (10/74)
Self-Confident
(10/165)
Self-Respectful
(11/195)
Coping (3/8)
Obsessive-Compulsive
(1/4)
Patient (11/547)
Temperamental
(11/264)
Emotional
Sensitivity
Neuroticism
Emotional (10/107)
Exaggerate (6/16)
Sensitive (11/179)
Complaining (9/47)
Content (6/26)
Depressive (3/14)
Neurotic (1/3)
Tense (4/10)
Example Response (Language)
Balanced person (English)
Quite calm, not rattled easily
(English)
He behaves like a young boy
(Tsonga)
She gets angry easily (Ndebele)
Is brave and is able to kill a
snake alone (Swati)
She gets easily scared (Tswana)
Craves attention (Afrikaans)
Difficult to please (Xhosa)
He is always needy and expects
others to sympathize with
him (Tsonga)
Believed in himself (Zulu)
Has self-respect (Venda)
Copes very well (English)
Obsessive behavior like tea in
the morning, then the pills
(English)
He does not get impatient with
you when you talk
to him; he would listen
to you attentively before
answering you
(N. Sotho)
One minute she is happy, the
next minute she is angry (N.
Sotho)
Cries a lot (Tswana)
Overreacted (English)
Easily gets hurt (Xhosa)
Real moaner; moans about
everything (Afrikaans)
She never gets satisfied
(N. Sotho)
Depressed (Afrikaans)
Neurotic (English)
Gets stressed out over small
things (Xhosa)
(Continued)
Personality Structure in South Africa
927
Table 2 (Cont.)
Cluster
Extraversion
Subcluster
Dominance
Facet
Assertive (11/238)
Authoritarian (11/350)
Disciplining (11/488)
Expressiveness
Strict (11/478)
Captivating (11/199)
Emotional Sharing
(11/345)
Noisy (9/109)
Outspoken (8/61)
Secretive (9/145)
Straightforward
(11/152)
Positive
Emotionality
Cheerful (11/810)
Humorous (11/704)
Sociability
Optimistic (9/87)
Playful (10/134)
Pleasure Seeking
(8/68)
Vivacious (10/175)
Communicative
(11/146)
Extravert/Introvert
(8/246)
Reserved (8/138)
Shy (11/190)
Sociable (11/1508)
Spontaneous
(2/67)
Storyteller (11/115)
Talkative (11/1239)
Example Response (Language)
Stand for her viewpoint
(Venda)
He wanted things to be done
his way (S. Sotho)
She liked to instill discipline
(Tsonga)
Strict and bossy (Tswana)
People just got drawn to her
(English)
If something has upset him,
he tells me (Zulu)
A noisy person (S. Sotho)
Outspoken, especially when
someone is wrong
(Xhosa)
He did not want to talk
about his past or future
(Tsonga)
He is a straightforward and
straight-talking person
(N. Sotho)
Always in a jovial mood; is
never in a bad mood (Swati)
He is full of jokes
(Ndebele)
Very positive (Venda)
A playful person (Tswana)
Likes to have fun (Xhosa)
Energetic (Afrikaans)
I love communicating with
people (S. Sotho)
She is an introvert (Swati)
He is reserved (Zulu)
Shy, but if you get to know me,
you would understand me
(English)
He enjoys being with people
(N. Sotho)
Spontaneous (Afrikaans)
She likes to tell about the times
when she was still a girl
(Zulu)
I like chatting with people
(Tsonga)
(Continued)
928
Nel, Valchev, Rothmann, et al.
Table 2 (Cont.)
Cluster
Facilitating
Subcluster
Encouraging
Others
Facet
Example Response (Language)
Aspirations for Others
(10/146)
Encouraging (11/694)
Wishes for everyone to succeed
(Xhosa)
Likes to encourage and
motivate people
(Swati)
Comes up with ideas, solutions,
and suggestions that make
you realize things (English)
Brings out the lighter side in me
(English)
Gives advice about life (Venda)
He taught me so many things
(Ndebele)
She is able to guide others
(Tsonga)
A person who inspires
(Tswana)
He is a leader at school and
in the community as well
(N. Sotho)
She is respected by people
in the village
(S. Sotho)
He is a role model to me (Zulu)
Discriminates, does not buy
clothes for everybody
(Swati)
Fair, not prejudiced (Afrikaans)
Honest (Xhosa)
Sound values and integrity
(English)
Loyal—to duties and as
friend (English)
He does not like people to
do bad things
(N. Sotho)
A person who pretends to
like you, whereas he does
not (Zulu)
He is responsible (Tswana)
Reliable and trustworthy
(Venda)
She likes telling the truth
(Ndebele)
Thought-Provoking
(1/3)
Uplifting (8/56)
Guidance
Advising (11/885)
Didactic (10/351)
Guiding (10/199)
Influential (6/68)
Leading (8/66)
Respectable (6/49)
Integrity
Fairness
Role Model (11/195)
Discriminative
(11/544)
Integrity
Fair (10/140)
Honest (11/420)
Integrity (3/17)
Loyal (5/110)
Morally Conscious
(9/459)
Pretending (11/160)
Responsible (11/403)
Trustworthy (11/1058)
Truthful (11/589)
(Continued)
Personality Structure in South Africa
929
Table 2 (Cont.)
Cluster
Intellect
Subcluster
Aesthetics
Facet
Example Response (Language)
Artistic (4/18)
Artistic and creative with lots
of interests (Afrikaans)
He loved handiwork
(Tsonga)
Creative, makes furniture
(English)
She is a good singer (Swati)
He has many talents
(Tswana)
Is able to see where the
problem lies (Xhosa)
He understands or knows
history well, and wild
animals (Ndebele)
Rational and logical
(English)
He understands himself
(S. Sotho)
He taught history nicely
and explained beautifully
(Zulu)
He does his work well (Tswana)
He is a person who owns and
runs his shop very well
(Zulu)
He is useless (Tsonga)
She could easily see when you
had a problem (Zulu)
Knows how to deal with people
(Xhosa)
He understands my traditions
(Venda)
Dreamer (Xhosa)
I am an independent-minded
person (S. Sotho)
Individualistic (Afrikaans)
He is interested in other
languages as well
(N. Sotho)
Prim and proper (English)
Conservative (Afrikaans)
I’m a religious person (Tsonga)
Liked traditional things (Swati)
Visionary (Xhosa)
Concrete Work
(10/265)
Creative (9/76)
Musical (5/15)
Talented (8/21)
Reasoning
Intelligent (10/443)
Knowledgeable
(11/105)
Logical (7/47)
Self-Insight (5/12)
Skillfulness
Articulative (11/206)
Competent (10/104)
Enterprising (11/139)
Openness
Social Intellect
Useless (3/5)
Perceptive (9/116)
BroadMindedness
Socially Intelligent
(6/20)
Understanding
(10/463)
Dreamer (4/13)
Independent (11/312)
Individualistic (3/25)
Open-Minded (10/127)
Prim and Proper (1/3)
Progressive (9/148)
Religious (11/1381)
Traditional (11/469)
Visionary (3/6)
(Continued)
930
Nel, Valchev, Rothmann, et al.
Table 2 (Cont.)
Cluster
Subcluster
Facet
Example Response (Language)
Epistemic
Curiosity
Academically Oriented
(11/240)
Eager to Learn
(11/209)
Inquisitive (11/153)
Materialism
Fashion-Conscious
(11/293)
Materialistic (11/86)
Adventurous (4/36)
Like to Travel (8/70)
Accommodating (5/26)
She likes to be educated
(Ndebele)
Likes to learn about other
people’s culture (Venda)
A person who likes to know the
answers of life (Tswana)
Is always well dressed in
current fashion (Swati)
Likes money (Zulu)
Adventurous (English)
She likes traveling (S. Sotho)
Addressed us in English so we
could understand (Xhosa)
She is approachable; I could
speak to her about anything
(S. Sotho)
He thinks he is better than all
the other people
(N. Sotho)
Flexible to situation (Tswana)
She is a humble and
down-to-earth person
(Ndebele)
Accepts people for who and
what they are (English)
Is proud and thinks of herself
better than others (Swati)
Was stubborn, did not listen to
anybody (Tswana)
Tolerant (Afrikaans)
Welcoming—to everyone
(Venda)
Likes to quarrel (Xhosa)
Openness to
Experience
Relationship
Harmony
Approachability
Approachable (11/311)
Arrogant (11/339)
Flexible (7/112)
Humble (11/247)
Open for Others (8/65)
Proud (11/126)
Stubborn (11/320)
Tolerant (7/34)
Welcoming (10/107)
Conflict
Seeking
Argumentative
(10/105)
Provoking (5/59)
Troublesome (11/337)
Interpersonal
Relatedness
Appeasing (9/37)
Constructive (6/37)
Cooperative (8/116)
Forgiving (10/159)
Provocative and calls people
names (Swati)
Creates tension for nothing
(Zulu)
If she made you angry, she will
come to your house and
apologize (N. Sotho)
Shares constructive ideas
(Xhosa)
Works well with others
(Tswana)
She holds no grudges (Tsonga)
(Continued)
Personality Structure in South Africa
931
Table 2 (Cont.)
Cluster
Subcluster
Facet
Example Response (Language)
Good Relations With
Another (10/529)
Peaceful (11/458)
Maintains a good relation with
others (Venda)
He likes peace among people
(N. Sotho)
He likes to bring peace where
there is misunderstanding
(Ndebele)
Doesn’t ask nicely (Afrikaans)
Peacekeeping (10/174)
Meddlesomeness
Well-Mannered
(11/648)
Gossiping (11/545)
Interfering (11/121)
Softheartedness
Active
Support
Community
Involvement
(11/143)
Heedful (11/426)
Helpful (11/1561)
Protective (9/46)
Solving Problems of
Others (11/159)
Supportive (11/618)
Amiability
Friendly (11/740)
Irritating (7/93)
Kind (11/1288)
Likable (10/183)
Pleasant (9/201)
Stern (7/24)
Egoism
Generous (11/1180)
Greedy (8/29)
Jealous (11/306)
Self-Centered (9/71)
Selfish (11/390)
A person who spreads rumors
about other people (S.
Sotho)
Likes to interfere in other
people’s business (English)
There is one person who is
always looking after the
community (Zulu)
She listens when you talk to her
(S. Sotho)
Is helpful when you are in need
(Swati)
Protective (Xhosa)
If I have a problem, she knows
how to solve it (Ndebele)
I like to give people my support
(Tswana)
She is a friendly person
(Tsonga)
He is annoying and irritating
(S. Sotho)
Kind (Venda)
He is loved by everyone (S.
Sotho)
He was a nice person to live
with (Zulu)
Always serious, not smiling
(Xhosa)
One who is generous and gives
food when asked (Swati)
Greedy (Afrikaans)
A person who is jealous of
other people’s possessions
(Zulu)
All revolves around her, she
thinks (English)
Wants everything for himself
(Xhosa)
(Continued)
932
Nel, Valchev, Rothmann, et al.
Table 2 (Cont.)
Cluster
Subcluster
Empathy
Facet
Agreeing (7/19)
Caring (11/1689)
Compassionate
(11/443)
Considerate (8/174)
Humane (6/52)
Loving (11/2903)
Respectful (11/1120)
Gratefulness
Satisfying Others
(3/10)
Appreciative (10/116)
Grateful (11/59)
Hostility
Abusive (11/293)
Aggressive (11/601)
Critical (10/159)
Cruel (11/475)
Delinquent (11/543)
Denigrating (10/326)
Distrustful (9/95)
Exploiting (10/79)
Intimidating (11/65)
Verbally Aggressive
(11/461)
Wrathful (1/11)
Example Response (Language)
Agreeable (Tswana)
Cares about other people
(English)
She feels pity for you when you
are in trouble (N. Sotho)
Considers others’ feelings
(Afrikaans)
He is good-natured and shows
humanity (Swati)
Loving and caring—concerned
about my life (Venda)
He respects other people
(Tsonga)
Makes people happy all the
time (Xhosa)
She doesn’t appreciate the good
of other people (Ndebele)
He is not thankful for what
people do for him
(N. Sotho)
Abusive—physically and
emotionally (English)
He is aggressive and likes
fighting (Tswana)
He likes criticizing others
(Tsonga)
He is a cruel person
(S. Sotho)
Mugged people (Xhosa)
Likes to belittle others (Venda)
He mistrusts people (English)
Uses other people (Afrikaans)
People were afraid of him
(N. Sotho)
Swears at his parents (Zulu)
Is wrathful and scolds,
especially when you have
disappointed her (Swati)
Note. The numbers in parentheses in the Facet column indicate the number of languages where the
facet appears and the number of responses represented under that facet.
of labeling, categorizing, and clustering was conducted mainly by the
principal author, but closely monitored and extensively discussed with the
other authors and members of the SAPI project. Personality-descriptive
terms were discussed in frequent group meetings with the collaborators of
this project in order to ensure adequacy and consistency of the analysis.
Personality Structure in South Africa
933
Several workshops were conducted at different stages of the analysis, in
which cultural and linguistic experts on the studied cultural groups provided feedback on the adequacy of the categorization and the ensuing
personality facets. The feedback from these workshops was taken into
account in the further modification of the conceptual clustering. Individual discussions were held with cultural and personality experts on the
final outcomes of the semantic clustering analysis, which allowed some
final refinements to the model to be made.
Results
In the following paragraphs, the nine clusters are presented in alphabetical order, with a brief description of their content (see Table 2 for
a full overview and examples of characteristic responses).
The Conscientiousness cluster represents an orientation toward
achieving things; having passion, determination, and perseverance
in the goals one sets for oneself; being precise and thorough, tidy,
punctual, careful and well organized, and caring about order; and
the ability to behave according to expectations. On the negative
pole, this cluster includes the characteristics of being forgetful and
reckless.
Emotional Stability refers to the emotional balance of a person,
the disposition to bravery and courage, the quality of being independent, confiding in one’s own abilities and having a positive view of
oneself, and the ability to control one’s emotions and their expression, as well as to handle challenging life situations. On the negative
pole, the cluster includes the tendency to be dissatisfied and complain, and proneness to depressive moods and stress.
The Extraversion cluster accounts for characteristics such as the
tendency to control others forcefully, being open to share or communicate with other people, being energetic and upbeat and seeing
the positive side of life, and the tendency to associate with others and
enjoy having people around oneself.
Facilitating represents the ability to guide others through life by
giving advice, teaching about right and wrong, and providing personal example as a role model, and the ability to motivate and
encourage others so they realize their potential.
Integrity refers to the quality of being honest, loyal, and reliable;
having principles and adhering to basic social norms of accepted
behavior; and the inclination to accept and treat all people equally,
rather than discriminate and favor some people over others.
934
Nel, Valchev, Rothmann, et al.
Intellect represents the quality of being creative and talented, the
capacity to attain insight in things in general and one’s self in particular; having knowledge and sharing it with others; the ability to do
things well, and the ability to understand others and social situations
and to react adequately.
Openness represents the quality of being receptive of different
ideas and appreciating progress, being eager to learn new things or
skills, the fondness of material possessions, and the inclination
toward traveling and seeing and experiencing new things.
Relationship Harmony encompasses characteristics such as being
approachable and accessible for others (vs. placing oneself above
others), being constructive in one’s relationships, and actively maintaining them by being forgiving, peaceful, and cooperative. On the
negative pole, the cluster includes the characteristics of being disruptive, causing (and enjoying) conflicts, and provoking others, as well
as interfering in others’ lives by gossiping or meddling.
Softheartedness represents the qualities of being pleasant and
kind, being concerned with the welfare of others, having appreciation of life and gratitude to others, having compassion, considering
other people’s needs and feelings and caring for them, and being
generous and actively involved with the well-being of one’s peers and
broader community. Subclusters from the negative pole are egoism
and hostility.
Relationship Harmony and Softheartedness are related; yet, the
two clusters have a different focus. Relationship Harmony refers
more to behaviors aimed at maintaining good relationships with
others, whereas Softheartedness deals more with nurturing and personal characteristics conducive for establishing or maintaining good
relationships, focusing less on the relationship itself.
Discussion
The aim of the present study was to explore the indigenous personality concepts of speakers of the 11 official languages in South Africa.
The 49,818 personality-descriptive responses from the semistructured interviews were condensed in successive steps to 188 facets, 37
subclusters, and nine broad clusters. The nine-cluster conceptual
model displays both similarities and differences with the dominant
personality models such as the Big Five. The Extraversion, Softheartedness, Conscientiousness, Emotional Stability, Intellect, and
Personality Structure in South Africa
935
Openness clusters broadly correspond to the respective Big Five
concepts. Softheartedness covers Agreeableness-related concepts,
while our Intellect and Openness clusters may be two components of
Openness in the FFM, where both labels for the factor have been
used (De Raad & Van Heck, 1994). It is notable that the Softheartedness cluster, with six subclusters and 39 facets, has the largest array
of personality concepts.
The remaining three clusters (Integrity, Relationship Harmony,
and Facilitating), on the other hand, seem to be less strongly related
to the Big Five model. Integrity has some similarity with the
Honesty factor of the HEXACO model (Ashton & Lee, 2001);
however, our cluster has a greater emphasis on issues of fairness and
discrimination. Relationship Harmony seems somewhat related to
the Interpersonal Relatedness construct of the CPAI (F. M. Cheung
et al., 2001). At the same time, it includes elements that are traditionally subsumed under Agreeableness (e.g., the Approachable and
Tolerant facets) and does not include elements of face saving, which
are characteristic of the CPAI’s Interpersonal Relatedness factor.
Finally, the Facilitating cluster is not covered in any of the Western
models of personality. It is instructive to consider the distinction of
this cluster from the Dominance subcluster (under Extraversion).
Dominance stands for being assertive and forceful, even using
intimidation or dictatorial tactics to acquire the compliance of
others. Facilitating, on the other hand, refers to the beneficial influence of a person on others; a person with this characteristic is well
respected and seen as a role model and a positive example for the
community.
Softheartedness and the three more culture-specific constructs
(Integrity, Relationship Harmony, and Facilitating) all refer to
aspects of social-relational functioning of a person. In this respect,
they can be considered as elaborations and extensions of aspects
that are represented by Agreeableness in the Big Five model (see,
e.g., Graziano & Eisenberg, 1997). The richness and density of
representations of social and relational aspects in the South African
implicit personality conceptions are an important finding of this
study.
Elements of Ubuntu (Nolte-Schamm, 2006) can also be recognized in clusters of the Agreeableness domain, especially Relationship Harmony and Softheartedness. These characteristics were
recognizable in more than one cluster and, importantly, they were
936
Nel, Valchev, Rothmann, et al.
recognizable in all languages. This indicates that Ubuntu concepts
may function as a fairly broad underlying frame of reference that
spans different personality clusters and cultural-linguistic groups in
South Africa.
STUDY 2
The process of condensing the original utterances to nine clusters was
done in several steps and accompanied by extensive consultation.
Still, an Achilles’ heel of this approach is its unknown validity.
Leaving the realm of qualitative methods, we wanted to address the
validity of a part of this process. We turned to a quantitative exploration of the higher level grouping of the 37 subclusters. We aimed to
estimate to what extent a grouping similar to the nine-cluster conceptual model would emerge when laypeople, who did not know our
final clustering, rated the relations among the 37 subclusters. We
were primarily interested in the replication of the model in South
Africa, where it had originated; however, we also employed a smallscale study in the Netherlands that could serve as a frame of reference to indicate possible cultural influences on the perceived relations
between the personality concepts.
Method
Participants
In South Africa, questionnaires were administered to 204 students at the
University of Johannesburg majoring in the social sciences. Participants’
ages ranged from 18 to 50 years (M = 21.57, SD = 4.69); 157 were females,
42 were males, and five persons did not indicate their gender. Forty-one
persons self-identified as White, 138 as Black, 14 as Coloured, and 5 as
Asian or Indian; 6 persons failed to answer the ethnicity question. The
sample included first-language speakers of Afrikaans (n = 2), English
(n = 20), Northern Sotho (n = 21), Southern Sotho (n = 13), Tswana
(n = 23), Ndebele (n = 4), Swati (n = 12), Xhosa (n = 12), Zulu (n = 20),
Tsonga (n = 12), Venda (n = 6), and other European (n = 3) and African
(n = 3) languages; data on first language were missing for 53 persons. The
students were not informed about the results of the conceptual cluster
analysis reported before.
In the Netherlands, questionnaires were administered to 95 social
science students at Tilburg University (77 females, 17 males, 1 unidentified) aged 18 to 32 years (M = 20.56, SD = 2.81). Participants were of
Personality Structure in South Africa
937
Dutch (n = 80), Turkish (n = 3), other European (n = 4), African (n = 1),
and South American (n = 1) origin; the ethnic origin of six persons was not
specified.
Instrument
The questionnaire was devised in English. It comprised a list of 666 pairs
of personality characteristics: The 37 subcluster labels were crossed, yielding 666 (= 37 ¥ 36 / 2) pairs.1 Brief descriptions of all characteristics, based
on the content of the subclusters as it emerged from the semantic analysis
(similar to the descriptions provided in the Results section of Study 1),
were provided, and participants were instructed to familiarize themselves
with them. For each pair, participants were asked to rate the extent to
which the two characteristics are related to each other. Participants were
instructed to rate the characteristics as related if they indicated either
similar (e.g., “love” and “devotion”) or opposite things (e.g., “love” and
“hatred”) but as unrelated if they indicated things that have nothing or
hardly anything to do with each other (e.g., “love” and “smartness”).
Relatedness was rated on a 5-point Likert scale, ranging from 1 (not at all
related) to 5 (very strongly related). The questionnaire and the descriptions
of the personality characteristics were translated into Dutch for the study
in the Netherlands using a committee approach. Completion of the questionnaire took one hour on average.
Results
We calculated the average scores for each item (characteristics pair)
across respondents. These scores were imputed in a symmetric matrix
of proximities between the individual characteristics (37 subclusters).
This matrix was subjected to a hierarchical cluster analysis using the
average-between-group-linkage method.
The outcomes for the South African data (see Figure 1) suggest
that on the highest level there is a distinction between positive
and negative characteristics. The positive characteristics seem to be
further divided into person-centered and relationship-centered
clusters. (A related interpretation would be in terms of agentic vs.
1. Study 2 was conducted at a point before two final refinements had been made
to the conceptual model presented in Study 1. As a result, there were two differences in the subclusters employed in Study 2: There was no Balance subcluster (its
facets being included under other subclusters), and there was a Politeness subcluster (including the Prim-and-Proper and Well-Mannered facets). These differences
between the versions of the conceptual model are minimal and do not substantially
restrict their comparability.
938
Nel, Valchev, Rothmann, et al.
Dendrogram Using Average Linkage (Between Groups)
Achievement
Orientation
Dedication
Self-Discipline
Orderliness
Aesthetics
Skillfulness
BroadMindedness
Openness
Epistemic
Curiosity
Courage
Ego-Strength
Emotional
Control
Emotional
Sensitivity
Fairness
Reasoning
Amiability
Politeness
Positive
Emotionality
Active Support
Encouraging
Others
Empathy
Guidance
Sociability
Social Intellect
Approachability
Interpersonal
Relatedness
Integrity
Expressiveness
Gratefulness
Conflict
Seeking
Meddlesomeness
Dominance
Egoism
Hostility
Thoughtlessness
Neuroticism
Materialism
Figure 1
Dendrogram of hierarchical cluster analysis on South African data.
Personality Structure in South Africa
939
communal characteristics, Bakan, 1966; what goes against it is the
fact that the Extraversion components, which usually are agentic
features, seem to reside under the communion/relationship-centered
grouping.) On a lower, more specific level, the following configurations emerge (see dotted line in the figure). A Conscientiousness
cluster emerges nearly identical to the conceptual model, except for
the Thoughtlessness element, which goes to the negative valence
supercluster. The same is true for the Openness cluster, with the
respective exception of Materialism. Two of the four Intellect characteristics come out in one cluster; differently from expectations,
Reasoning forms a cluster with Fairness, and Social Intellect goes to
the cluster of Relations and Social Functioning. Four of the five
Emotional Stability characteristics form two clusters that are separate yet close to each other; Neuroticism goes to the negative valence
supercluster.
The broad cluster of relations and social functioning accommodates elements of proper Agreeableness (Amiability, Politeness, and
Positive Emotionality), caring and guiding (Active Support, Encouraging Others, Empathy, and Guidance), social-relational orientation
(Sociability, Social Intellect, Approachability, and Interpersonal
Relatedness), and Integrity (Fairness, Reasoning, and Integrity).
These elements broadly represent the concepts of Extraversion, Softheartedness, Integrity, Relationship Harmony, and Facilitating from
the conceptual model. The structure of the conceptual model is not
replicated exactly, but there are some marked correspondences; for
instance, the concepts of Empathy and Active Support, Guidance,
and Encouraging Others seem to be related as expected in the context
of the Softheartedness and Facilitating clusters (whereby the two
clusters might, in turn, have a strong relation).
Finally, the negative valence supercluster accommodates all negative elements from different conceptual clusters. It is worth noting
that even within this supercluster, groupings conform to the expectations from the conceptual model: Conflict and Meddlesomeness,
on the one hand, and Egoism and Hostility, on the other hand,
represent the negative poles of Relationship Harmony and Softheartedness, respectively.
The outcomes of the Dutch data (Figure 2) are fairly similar to the
South African data. Conscientiousness, Openness, and, to a lesser
extent, Intellect and Emotional Stability are clearly distinguishable
as stand-alone clusters. Most negative concepts tend to group
940
Nel, Valchev, Rothmann, et al.
Dendrogram Using Average Linkage (Between Groups)
Epistemic
Curiosity
Openness
BroadMindedness
Aesthetics
Reasoning
Skillfulness
Achievement
Orientation
Dedication
Orderliness
Self-Discipline
Courage
Expressiveness
Emotional
Sensitivity
Emotional
Control
Integrity
Fairness
Politeness
Amiability
Gratefulness
Encouraging
Others
Active Support
Empathy
Interpersonal
Relatedness
Social Intellect
Positive
Emotionality
Sociability
Approachability
Conflict Seeking
Hostility
Dominance
Guidance
Meddlesomeness
Egoism
Ego-Strength
Materialism
Thoughtlessness
Neuroticism
Figure 2
Dendrogram of hierarchical cluster analysis on Dutch data.
Personality Structure in South Africa
941
together, and there is a large cluster accommodating social-relational
concepts. Differently from the South African data, Facilitating fails
to emerge as a grouping within the social-relational cluster and there
is a weaker relation between the two negative aspects of Relationship Harmony (Conflict Seeking and Meddlesomeness); on the other
hand, Integrity emerges as a distinguishable grouping within the
social-relational cluster and so do two elements of Extraversion
(Positive Emotionality and Sociability).
Discussion
The outcomes of the hierarchical cluster analysis are to some extent
close to the structure of the conceptual model that was derived in the
qualitative analysis. Conscientiousness, Openness, Intellect, Emotional Stability, and Facilitating are easily recognizable as clusters.
Extraversion, Softheartedness, Integrity, and Relationship Harmony
are less clearly distinguished within the broad cluster of relations and
social functioning.
The overall division in negative and positive characteristics, which
accounts for many of the discrepancies between the hierarchical
cluster analysis and the conceptual model, is in agreement with findings from the lexical literature (Saucier & Goldberg, 2001). Whereas
in the perceptions of laypeople the differentiation of positive from
negative characteristics is apparently the most important grouping
factor, our conceptual analysis of the qualitative data has focused on
the content of and relations among the subclusters, independent of
their valence.
The second source of discrepancies refers to the emergence of one
global cluster of relations and social functioning, where the fine
distinctions between the concepts do not appear to be clearly drawn
in the perceptions of laypeople. Extraversion, Softheartedness,
Integrity, Relationship Harmony, and Facilitating are to some extent
intertwined. Nonetheless, several of their elements clearly group
together as expected. This broad social cluster also attracts Social
Intellect, indicating that in the perceptions of participants, this
concept is primarily important for its social, rather than intellectual,
functions. The failure to replicate the finer distinctions in the interpersonal domain may to some extent be attributable to the relatively
high demands of the similarity judgment task, involving 666 pairwise comparisons.
942
Nel, Valchev, Rothmann, et al.
The Dutch data as a whole demonstrate important similarities
with both the South African data and the conceptual model. In
addition, we found some differences between the Dutch and South
African data that seem informative on the cultural meaning of the
derived personality constructs. Facilitating is more readily recognized as a personality concept in South Africa than in the Netherlands; in this sense it may indeed represent an indigenous personality
concept. In a similar manner but in the opposite direction, the relation between the constituting elements of Extraversion (Positive
Emotionality and Sociability) and Integrity (Integrity and Fairness)
seems to be more salient in the conceptions of Dutch people than
South Africans. The latter observation may imply that to some
extent, our conceptual model has inadvertently been influenced somewhat by our own Western (theoretical) perspectives on personality.
In conclusion, quantitative data on the perceived relations
between the 37 midlevel subclusters provide general support for the
adequacy of the qualitative clustering of the first study, although
several of the conceptual clusters in the domain of relations and
social functioning did not replicate in detail. This incomplete overlap
of the findings of the two studies points to the necessity to validate
the structure in a more elaborate way by administering items derived
from the clusters to representative samples of various ethnic groups
in South Africa. This study will clarify whether the two related
clusters in the social domain, Relationship Harmony and Softheartedness, are distinct as observed in the first study or are more likely to
merge in a social supercluster as found in the second study. Such a
study is currently underway.
GENERAL DISCUSSION
We set out to explore the implicit personality structure in South
Africa’s 11 official languages. Rather than starting from existing
personality models, we employed an indigenous approach, in which
the implicit personality structure is derived from everyday conceptions of personality. In the first study, we obtained personality
descriptions by means of semistructured interviews in samples of
speakers of each of the 11 languages. In consecutive steps of semantic
clustering and conceptual analysis of these personality descriptions,
we formed nine broad clusters of personality concepts. In the second
Personality Structure in South Africa
943
study, we employed a quantitative analysis of the midlevel components of these clusters in two independent samples. This analysis
provided general support for the model, although some elements
were not replicated in detail.
The nine-cluster model displays a certain correspondence with
established models of personality like the Big Five and HEXACO.
Our findings do not contradict claims of universality of personality
dimensions of these models (see, e.g., Church, 2008). At the same
time, our model differs from these established models in two ways.
Firstly, the Agreeableness-like cluster, Softheartedness, is considerably larger than the rest. This finding has some relation to findings in
the lexical studies, where the general tendency is for Extraversion and
Agreeableness to be the largest factors (De Raad et al., 2010; John,
Naumann, & Soto, 2008; Saucier & Goldberg, 2001). In our study,
however, the concepts related to Agreeableness and social-relational
functioning (represented, besides Softheartedness, in Relationship
Harmony, Integrity, and Facilitating) outnumber the rest, including
Extraversion, in an impressive manner. It is also worth noting that
Extraversion is a relatively narrow cluster in our data as compared to
most lexical studies, in which aspects of confidence and boldness are
often more salient (Peabody & De Raad, 2002). Secondly, three of
the clusters are relatively foreign to the Big Five model. Integrity and
Relationship Harmony are reminiscent of the HEXACO model’s
Honesty (Ashton & Lee, 2001; Lee & Ashton, 2008) and the CPAI’s
Interpersonal Relatedness (F. M. Cheung et al., 2001), respectively;
yet the clusters have somewhat different connotations in our data.
Our Integrity cluster has a correspondence to the first two personality facets of the Honesty-Humility factor, sincerity and fairness, but
not to the other two, greed-avoidance and modesty. Additionally,
Integrity includes facets associated with equal treatment (vs. discrimination), which is not represented in the HEXACO model. The
CPAI’s Interpersonal Relatedness, in turn, consists of four core components: harmony, ren qing (relationship orientation), flexibility, and
face (F. M. Cheung et al., 2001). Of those, the first three, representing respectively intra- and interpersonal harmony, active efforts
to maintain harmony, and flexibility to situations, seem related to
our Relationship Harmony cluster. Face (referring to face-saving
motives) seems to be less salient in the studied languages.
Finally, Facilitating, referring to the qualities of an individual as
a good guide in life and example to others, seems to be a fairly unique
944
Nel, Valchev, Rothmann, et al.
concept that is not represented in any personality model (although
see De Raad, 1999). It could be argued that the Facilitating cluster
was relatively salient in our data because we asked informants to
describe persons who often serve a facilitating role in the socialization process, such as teachers. However, many responses that were
coded as belonging to the Facilitating cluster were found in the
descriptions of persons who are not typically associated with this
role, such as siblings and friends.
Both aspects in which our model differs from the Big Five—the
overrepresentation of relational, Agreeableness-like concepts and
the identification of concepts not well represented in the Big Five—
point in the same direction. The attributes of an individual’s
social-relational functioning seem to warrant a central place in the
personality conceptions of South Africans, to such an extent that
the Big Five conceptual space has to be expanded to accommodate
these attributes. It is noteworthy that the strongest claims for expansion of the Big Five, coming from China (Church, 2008; F. M.
Cheung et al., 2001), also refer to social-relational factors. The collectivistic values of a culture (Hofstede, 1980) can be expected to be
associated with an emphasis on relational aspects of personality,
although indigenous research in other collectivistic cultures such as
Mexico (Ortiz et al., 2007) and the Philippines (Katigbak et al., 2002)
has found less support for culture-specific dimensions beyond the Big
Five. Interestingly, other research involving student samples from
Mexico and the Philippines (Del Prado et al., 2007) has also failed to
confirm hypotheses derived from the individualism–collectivism
theoretical framework for these two cultures, leaving the possibility
open that these cultures, or especially student samples there, may be
somewhat atypical with respect to characteristics of collectivism and
interdependence. It remains to be established in direct comparisons
of measures based on the present model and Chinese inventories to
what extent there is an overlap in their conceptual space in different
samples.
An important characteristic of our indigenously derived model of
personality conceptions in South Africa is that it represents data
from all 11 major cultural-linguistic groups of the country. The
model incorporates both common facets found across all or most of
the groups (which is true for the majority of the facets) and facets
found in only a few or single groups. In this way, the model accounts
comprehensively for the implicit structure of personality in all
Personality Structure in South Africa
945
groups, rather than favoring some groups over others. In other
words, the conceptual model presented in this study suggests a
derived-etic structure (Berry, 1989), which is the case when a psychological phenomenon is shown to be invariant across cultural groups,
using culture-specific methods. The structure accommodates the core
elements of personality deemed important in the different culturallinguistic groups of South Africa. The extent to which groups differ
in their perceptions of the specific composition of these core elements
(e.g., what makes up Intellect?) should be addressed in a future study.
The anticipated influence of the notion of Ubuntu was evident across
different clusters in the social-relational domain and in all culturallinguistic groups. The model developed in this study thus forms a
strong basis for the development of an instrument for the culturally
appropriate assessment of personality in South Africa.
Our study has implications for the emic-etic debate. After decades
of often ideological debates between proponents of both types of
studies, psychological research is now more receptive of rapprochement. Emic and etic studies should inform each other about more
universal and more culture-specific models of personality (F. M.
Cheung et al., 2011). Thus, on the one hand, our clustering of emic
terms was partly informed by current, typically etic models in personality such as the FFM and HEXACO model. On the other hand,
clusters that are found in South Africa (notably Relationship
Harmony and Softheartedness) may have at least some applicability
in other cultural contexts. So emic approaches may inform etic
approaches as to how their models could be expanded. The final goal
of the combination of emic and etic approaches is not a classification
of purely universal and purely culture-specific aspects of personality
but a better appreciation of which aspects are shared across which
types of cultures. The combination of emic and etic studies can help
to overcome the dichotomous view of personality traits as either
culture-specific or universal and give way to a more gradual view of
levels of universality and cultural specificity of traits.
REFERENCES
Abrahams, F., & Mauer, K. F. (1999). The comparability of the constructs of the
16PF in the South African context. Journal of Industrial Psychology, 25, 53–59.
Allport, G. W., & Odbert, H. S. (1936). Trait-names: A psycho-lexical study.
Psychological Monographs, 47(211).
946
Nel, Valchev, Rothmann, et al.
Ashton, M. C., & Lee, K. (2001). A theoretical basis for the major dimensions of
personality. European Journal of Personality, 15, 327–353.
Bakan, D. (1966). The duality of human existence: Isolation and communion in
Western man. Boston: Beacon Press.
Berry, J. W. (1989). Imposed etics-emics-derived etics: The operationalization of a
compelling idea. International Journal of Psychology, 24, 721–735
Berry, J. W., Poortinga, Y. H., Segall, M. H., & Dasen, P. R. (2002). Cross-cultural
psychology. Research and applications (2nd ed.). Cambridge, UK: Cambridge
University Press.
Cheung, F. M., Leung, K., Zhang, J. X., Sun, H. F., Gan, Y. G., Song, W. Z.,
et al. (2001). Indigenous Chinese personality constructs: Is the five-factor
model complete? Journal of Cross-Cultural Psychology, 32, 407–433.
Cheung, F. M., Van de Vijver, F. J. R., & Leong, F. T. L. (2011). Toward a new
approach to the assessment of personality in culture. American Psychologist,
66, 593–603.
Cheung, S. F., Cheung, F. M., Howard, R., & Lim, Y.-H. (2006). Personality
across the ethnic divide in Singapore: Are “Chinese traits” uniquely Chinese?
Personality and Individual Differences, 41, 467–477.
Church, A. T. (2001). Personality measurement in cross-cultural perspective.
Journal of Personality, 69, 979–1006.
Church, A. T. (2008). Current controversies in the study of personality across
cultures. Social and Personality Psychology Compass, 2, 1930–1951.
De Raad, B. (1999). Interpersonal lexicon: Structural evidence from two independently constructed verb-based taxonomies. European Journal of Psychological
Assessment, 15, 181–195.
De Raad, B., Barelds, D. P. H., Levert, E., Ostendorf, F., Mlacic, B., Di Blas, L.,
et al. (2010). Only three factors of personality description are fully replicable
across languages: A comparison of 14 trait taxonomies. Journal of Personality
and Social Psychology, 98, 160–173.
De Raad, B., & Van Heck, G. (Eds.). (1994). The fifth of the Big Five. European
Journal of Personality, 8, 225–356.
Del Prado, A. M., Church, A. T., Katigbak, M. S., Miramontes, L. G., Whitty, M.
T., Curtis, G. J., et al. (2007). Culture, method, and the content of selfconcepts: Testing trait, individual–self-primacy, and cultural psychology
perspectives. Journal of Research in Personality, 41, 1119–1160.
Foxcroft, C. D., Paterson, H., Le Roux, N., & Herbst, D. (2004). Psychological
assessment in South Africa: A needs analysis. Pretoria, South Africa: Human
Sciences Research Council.
Graziano, W. G., & Eisenberg, N. (1997). Agreeableness: A dimension of personality. In R. Hogan, J. A. Johnson, & S. R. Briggs (Eds.), Handbook of personality psychology (pp. 795–824). San Diego, CA: Academic Press.
Harkness, S., Moscardino, U., Bermudez, M. R., Zylicz, P. O., Welles-Nyström,
B., Blom, M., et al. (2006). Mixed methods in international collaborative
research: The experiences of the International Study of Parents, Children, and
Schools. Cross-Cultural Research, 40, 65–82.
Hofstede, G. (1980). Culture’s consequences: International differences in workrelated values. Beverly Hills, CA: Sage.
Personality Structure in South Africa
947
John, O. P. (1990). The “Big Five” factor taxonomy: Dimensions of personality in
the natural language and in questionnaires. In L. A. Pervin (Ed.), Handbook of
personality theory and research (pp. 66–100). New York: Guilford Press.
John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift to the integrative Big Five taxonomy: History, measurement, and conceptual issues. In O. P.
John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory
and research (3rd ed., pp. 114–158). New York: Guilford Press.
Katigbak, M. S., Church, A. T., Guanzon-Lapeña, M. A., Carlota, A. J., & del
Pilar, G. H. (2002). Are indigenous dimensions culture-specific? Philippine
inventories and the five-factor model. Journal of Personality and Social Psychology, 82, 89–101.
Kohnstamm, G. A., Halverson, C. F., Jr., Mervielde, I., & Havill, V. L. (Eds.).
(1998). Parental descriptions of child personality: Developmental antecedents of
the Big Five? Mahwah, NJ: Erlbaum.
Lee, K., & Ashton, M. C. (2008). The HEXACO personality factors in the indigenous personality lexicons of English and 11 other languages. Journal of
Personality, 76, 1001–1053.
Lin, E. J.-L., & Church, A. T. (2004). Are indigenous Chinese personality dimensions culture-specific? An investigation of the Chinese Personality Assessment
Inventory in Chinese American and European American samples. Journal of
Cross-Cultural Psychology, 35, 586–605.
McCrae, R. R., & Allik, J. (Eds.). (2002). The five-factor model of personality
across cultures. New York: Kluwer Academic/Plenum.
McCrae, R. R., Terracciano, A., & 78 Members of the Personality Profiles of
Cultures Project. (2005). Universal features of personality traits from the
observer’s perspective: Data from 50 cultures. Journal of Personality and Social
Psychology, 88, 547–561.
Meiring, D., Van de Vijver, F. J. R., & Rothmann, S. (2006). Bias in an adapted
version of the 15FQ+ in South Africa. South African Journal of Psychology, 36,
340–356.
Meiring, D., Van de Vijver, A. J. R., Rothmann, S., & Barrick, M. R. (2005).
Construct, item, and method bias of cognitive and personality measures in
South Africa. South African Journal of Industrial Psychology, 31, 1–8.
Mervielde, I. (1998). Validity of results obtained by analyzing free personality
descriptions. In G. A. Kohnstamm, C. F. Halverson Jr., I. Mervielde, & V. L.
Havill (Eds.), Parental descriptions of child personality: Developmental antecedents of the Big Five? (pp. 189–204). Mahwah, NJ: Erlbaum.
Nolte-Schamm, C. (2006). The African anthropology as resource for reconciliation: Ubuntu/Botho as a reconciliatory paradigm in South Africa. Scriptura,
93, 370–383.
Ortiz, F. A., Church, A. T., Vargas-Flores, J. D., Ibáñez-Reyes, J., Flores-Galaz,
M., Iuit-Briceño, J. I., et al. (2007). Are indigenous personality dimensions
culture-specific? Mexican inventories and the five-factor model. Journal of
Research in Personality, 41, 618–649.
Peabody, D., & De Raad, B. (2002). The substantive nature of psycholexical
personality factors: A comparison across languages. Journal of Personality and
Social Psychology, 83, 983–997.
948
Nel, Valchev, Rothmann, et al.
Ramsay, L. J., Taylor, N., De Bruin, G. P., & Meiring, D. (2008). The Big Five
personality factors at work: A South African validation study. In J. Deller
(Ed.), Research contributions to personality at work (pp. 99–114). Munich,
Germany: Rainer Hampp Verlag.
Saucier, G. (2009). Recurrent personality dimensions in inclusive lexical studies:
Indications for a big six structure. Journal of Personality, 77, 1577–1614.
Saucier, G., & Goldberg, L. R. (2001). Lexical studies of indigenous personality
factors: Premises, products, and prospects. Journal of Personality, 69, 847–879.
Statistics South Africa. (2001). Census 2001: Key results. Retrieved May 14, 2012,
from http://www.statssa.gov.za
Taylor, N., & De Bruin, G. P. (2005). Basic Traits Inventory: Technical manual.
Johannesburg, South Africa: Jopie van Rooyen & Partners SA.
Valchev, V. H., Van de Vijver, F. J. R., Nel, J. A., Rothmann, S., Meiring, D., &
De Bruin, G. P. (2011). Implicit personality conceptions of the Nguni culturallinguistic groups of South Africa. Cross-Cultural Research, 45, 235–266.